[SPARK-17069] Expose spark.range() as table-valued function in SQL

## What changes were proposed in this pull request?

This adds analyzer rules for resolving table-valued functions, and adds one builtin implementation for range(). The arguments for range() are the same as those of `spark.range()`.

## How was this patch tested?

Unit tests.

cc hvanhovell

Author: Eric Liang <ekl@databricks.com>

Closes #14656 from ericl/sc-4309.

2 files changed, 107 insertions, 0 deletions

diff --git a/sql/core/src/test/resources/sql-tests/inputs/table-valued-functions.sql b/sql/core/src/test/resources/sql-tests/inputs/table-valued-functions.sql
new file mode 100644
index 0000000000..2e6dcd538b
--- /dev/null
+++ b/sql/core/src/test/resources/sql-tests/inputs/table-valued-functions.sql
@@ -0,0 +1,20 @@
+-- unresolved function
+select * from dummy(3);
+
+-- range call with end
+select * from range(6 + cos(3));
+
+-- range call with start and end
+select * from range(5, 10);
+
+-- range call with step
+select * from range(0, 10, 2);
+
+-- range call with numPartitions
+select * from range(0, 10, 1, 200);
+
+-- range call error
+select * from range(1, 1, 1, 1, 1);
+
+-- range call with null
+select * from range(1, null);
diff --git a/sql/core/src/test/resources/sql-tests/results/table-valued-functions.sql.out b/sql/core/src/test/resources/sql-tests/results/table-valued-functions.sql.out
new file mode 100644
index 0000000000..d769bcef0a
--- /dev/null
+++ b/sql/core/src/test/resources/sql-tests/results/table-valued-functions.sql.out
@@ -0,0 +1,87 @@
+-- Automatically generated by SQLQueryTestSuite
+-- Number of queries: 7
+
+
+-- !query 0
+select * from dummy(3)
+-- !query 0 schema
+struct<>
+-- !query 0 output
+org.apache.spark.sql.AnalysisException
+could not resolve `dummy` to a table-valued function; line 1 pos 14
+
+
+-- !query 1
+select * from range(6 + cos(3))
+-- !query 1 schema
+struct<id:bigint>
+-- !query 1 output
+0
+1
+2
+3
+4
+
+
+-- !query 2
+select * from range(5, 10)
+-- !query 2 schema
+struct<id:bigint>
+-- !query 2 output
+5
+6
+7
+8
+9
+
+
+-- !query 3
+select * from range(0, 10, 2)
+-- !query 3 schema
+struct<id:bigint>
+-- !query 3 output
+0
+2
+4
+6
+8
+
+
+-- !query 4
+select * from range(0, 10, 1, 200)
+-- !query 4 schema
+struct<id:bigint>
+-- !query 4 output
+0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+
+
+-- !query 5
+select * from range(1, 1, 1, 1, 1)
+-- !query 5 schema
+struct<>
+-- !query 5 output
+org.apache.spark.sql.AnalysisException
+error: table-valued function range with alternatives:
+ (end: long)
+ (start: long, end: long)
+ (start: long, end: long, step: long)
+ (start: long, end: long, step: long, numPartitions: integer)
+cannot be applied to: (integer, integer, integer, integer, integer); line 1 pos 14
+
+
+-- !query 6
+select * from range(1, null)
+-- !query 6 schema
+struct<>
+-- !query 6 output
+java.lang.IllegalArgumentException
+Invalid arguments for resolved function: 1, null